To evaluate the extent to which relationships between apolipoprotein E, cognitive functioning, and survival in people aged 60 to 80 persist into advanced old age. Examine the effect of apolipoprotein E genotypes on baseline cognitive functioning, cognitive decline over 5 years, and survival in a cohort of 1,551 nonagenarians. The Danish 1905 birth cohort. One thousand five hundred fifty-one nonagenarians from the Danish 1905 birth cohort. Cognitive functioning was assessed using the Mini-Mental State Examination (MMSE) and five brief cognitive tests (cognitive composite). The subjects were stratified into four groups by occurrence of a protective (epsilon2) or a risk (epsilon4) apo E allele (epsilon22 and epsilon23, epsilon33, epsilon24 and epsilon34, epsilon44). At intake, the mean scores for the three genotype groups were 22.1, 21.8, 21.4, and 21.0 for MMSE and 0.10, 0.07, -0.02, and 0.30 for the cognitive composite, respectively. Growth-curve analyses showed that, although individuals carrying at least one epsilon4 allele had slightly lower MMSE scores and declined slightly more rapidly over time, this effect was not statistically significant and was not apparent in scores on the cognitive composite. In subjects whose functioning was relatively well preserved (those still living and able to participate in the assessment, and whose cognitive functioning had declined less than 4 points on the MMSE), epsilon4 frequencies tended to decline at subsequent waves (P=.03, chi-square test for trend), but epsilon4 had no significant survival disadvantage (hazard ratio=1.11 (95% confidence interval=0.99-1.25; P=.07). Apo E genotype has a small effect on the probability of remaining a well-functioning nonagenarian but no separately detectable effect on cognitive functioning, cognitive decline, or survival.
